基于13元组模型的运动人体时空关系描述方法  

The Description Method of Motion human body Temporal-spatial Relationship Based on 13 tuple temporal-spatial mode

作　　者：李敏[1,2] 陈燕平[1] 宋曰聪[1] 

机构地区：[1]绵阳师范学院信息工程学院,四川绵阳621006 [2]中国工程物理研究院五所,四川绵阳621000

出　　处：《绵阳师范学院学报》2017年第5期73-79,共7页Journal of Mianyang Teachers＇ College

基　　金：四川省教育厅资助项目(14ZA0257;142A0260);绵阳师范学院资助项目(2011A13;2013A10)

摘　　要：针对运动人体时空关系描述存在的问题,提出了基于13元组运动人体时空模型的运动人体时空关系描述方法 .该模型包含了二维到三维空间对运动人体的拓扑关系、方向关系、和时间关系.三维可视域和云模型的采用,实现了三维空间中不确定方向关系的判定.实验结果显示:该方法能全方位从点、线、面、体在二维至三维空间中实现视频场景中运动人体时空关系的自然语言描述.To solve the problem of the description of the spatial and temporal relationship of the human body, a method based on the 13 body motion model is proposed. The model includes the topological relation, direction relation and time relation of the moving human body from two dimensions to three dimensions. Three dimensional visibility and cloud model are used to determine the direction of uncertainty in three - dimensional space. The ex- perimental results show that this method can realize the natural language description of the motion of human body in the video scene from point, line and surface to body.

关 键 词：运动人体时空关系 13元组时空模型 自然语言描述 三维时空可视域 云模型 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]